Qwen3 Coder 480B A35B: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Alibabaalibaba/opensource | No data | — | 99.99% | 99.96%◆ (best) | 99.75% | 1.21s | 11.7s | 36 tps | 262K | $0.975 / $4.88 |
| Down | — | 98.33% | 96.04% | 96.02% | 617ms◆ (best) | 19.3s | 68 tps◆ (best) | 262K | $0.300 / $1.00 | |
Googlegoogle-vertex/us-south1 | No data | — | 99.96% | 99.85% | 99.90%◆ (best) | 640ms | 3.29s◆ (best) | 57 tps | 262K | $0.220 / $1.80◆ (best) |
| No data | — | 94.73% | 64.54% | 84.86% | — | — | — | 262K | $0.380 / $1.55 | |
| No data | — | 96.86% | 90.10% | 89.45% | 1.30s | 23.8s | 45 tps | 256K | $0.350 / $1.50 |
- No data24h 99.99%
- 7d
- 99.96%◆ (best)
- 30d
- 99.75%
- p50
- 1.21s
- p99
- 11.7s
- Throughput
- 36 tps
- Context
- 262K
- $/Mtok
- $0.975 / $4.88
alibaba/opensource - Down24h 98.33%
- 7d
- 96.04%
- 30d
- 96.02%
- p50
- 617ms◆ (best)
- p99
- 19.3s
- Throughput
- 68 tps◆ (best)
- Context
- 262K
- $/Mtok
- $0.300 / $1.00
deepinfra/turbofp4 - No data24h 99.96%
- 7d
- 99.85%
- 30d
- 99.90%◆ (best)
- p50
- 640ms
- p99
- 3.29s◆ (best)
- Throughput
- 57 tps
- Context
- 262K
- $/Mtok
- $0.220 / $1.80◆ (best)
google-vertex/us-south1 - No data24h 94.73%
- 7d
- 64.54%
- 30d
- 84.86%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.380 / $1.55
novita/fp8fp8 - No data24h 96.86%
- 7d
- 90.10%
- 30d
- 89.45%
- p50
- 1.30s
- p99
- 23.8s
- Throughput
- 45 tps
- Context
- 256K
- $/Mtok
- $0.350 / $1.50
venice/fp8fp8
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.
Price vs latency
One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag
History
Availability
Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.
Time to first token
Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window, averaged over the hour.
Throughput
Median tokens per second.